作者: Seyed Ahmad Mireei , Morteza Sadeghi
DOI: 10.1016/J.JFOODENG.2012.08.032
关键词: Linear discriminant analysis 、 Principal component analysis 、 Date Fruit 、 Chemistry 、 Ripening 、 Pattern recognition 、 Analytical chemistry 、 Classification methods 、 Artificial intelligence 、 Near-infrared spectroscopy 、 Data set 、 Partial least squares regression
摘要: Abstract This study introduces the application of near infrared spectroscopy (NIRs) to detect bunch withering disorder in date fruit (cv. Mazafati). The samples included intact as well infected fruits at different stages ripening. Chemometric evaluation data was performed by soft independent modeling class analogy (SIMCA), partial least squares discriminant analysis (PLS-DA), and principal components combined with artificial neural networks (PCA–ANN). PLS-DA algorithm able provide models best classification performance, followed SIMCA then PCA–ANN. maturity stage influenced performance methods. accuracy for late harvested better than those normal time set all analyses. total accuracies 82%, 93% 86%, respectively normal, sets demonstrate that NIRs has a strong potential fruit.